|
|
--- |
|
|
license: apache-2.0 |
|
|
language: |
|
|
- en |
|
|
tags: |
|
|
- medical |
|
|
- icd10 |
|
|
- medical coding |
|
|
- clinical |
|
|
- healthcare |
|
|
- icd10dx |
|
|
- CAC |
|
|
widget: |
|
|
- text: >- |
|
|
Chief Complaint: John has been experiencing shortness of breath and mild |
|
|
chest pain for the past two days. |
|
|
|
|
|
Medical History: He has a history of hypertension, which is currently under |
|
|
control with medication. Additionally, he was diagnosed with Type 2 Diabetes |
|
|
five years ago and had an episode of bronchitis last year. |
|
|
|
|
|
Examination Findings: During the examination, it was noted that Johns heart |
|
|
rate was elevated, and his blood pressure was slightly high at 150/95 mmHg. |
|
|
His respiratory rate was above normal limits, and his oxygen saturation was |
|
|
94% on room air. |
|
|
|
|
|
Investigations: An ECG was performed, which thankfully did not show any |
|
|
significant ST changes, typically indicative of acute coronary syndrome. A |
|
|
chest X-ray was also done, revealing no acute cardiopulmonary processes. |
|
|
|
|
|
Assessment and Plan: Given the elevated heart rate and hypertension, coupled |
|
|
with his shortness of breath, there is a concern for the early signs of |
|
|
congestive heart failure. However, the absence of acute changes in the ECG |
|
|
and the normal chest X-ray findings are reassuring. It is important to |
|
|
consider his history of hypertension and diabetes in the overall assessment, |
|
|
as these factors elevate his risk for cardiovascular complications. |
|
|
|
|
|
A conservative approach is recommended at this stage. We will start diuretic |
|
|
therapy to manage potential fluid overload and schedule an echocardiogram to |
|
|
further assess his cardiac function. In the meantime, his blood pressure and |
|
|
blood glucose levels will be closely monitored. A follow-up appointment is |
|
|
scheduled for next week, with instructions to return earlier if his symptoms |
|
|
worsen. |
|
|
example_title: Example 1 |
|
|
- text: 'Impression: fever, chills, cough, chest pain, shortness of breath, N/V.' |
|
|
example_title: Example 2 |
|
|
--- |
|
|
# ICD-10 DX Code Identification Model |
|
|
|
|
|
## Overview |
|
|
This model is designed for the identification of tokens related to ICD-10 DX codes in clinical documents. We focus on a subset of approximately 4,000+ codes, |
|
|
which are the most frequently used in clinical documentation. Please refer config.json file for target codes we used to train this model. |
|
|
|
|
|
## Model Details |
|
|
- **Type**: Named Entity Recognition (NER) |
|
|
- **Target**: ICD-10 DX Codes |
|
|
- **Code Subset**: 4,000+ most common codes |
|
|
|
|
|
## Dataset |
|
|
The dataset comprises clinical documents annotated for ICD-10 DX codes. We ensure a balanced representation of the selected codes to prevent model bias. |
|
|
the dataset is private one, used internally to trian the model. |
|
|
|
|
|
## Training |
|
|
Due to GPU memory constraints, training is conducted in epochs with periodic evaluations to monitor performance and mitigate overfitting. |
|
|
|
|
|
# Use a pipeline as a high-level helper |
|
|
|
|
|
from transformers import pipeline |
|
|
|
|
|
pipe = pipeline("token-classification", model="imperiumhf/imp_clinical_dxcode_ner_v2") |
|
|
|
|
|
## Evaluation |
|
|
Need to update metrics |
|
|
|
|
|
## Limitations and Considerations |
|
|
- Overfitting risk due to repeated training on the same dataset. |
|
|
- The balance between model complexity and the large number of classes. |
|
|
- Regular model evaluation for performance monitoring. |
|
|
|
|
|
## Contact |
|
|
krishnareddyn@kpmd.biz |
|
|
|
|
|
## Acknowledgements |
|
|
All the rights over this model is reserved for Imperium software solutions pvt ltd. |
